Product market competition and labour market rigidities: Firm-level evidence from South Africa

by Fransesco Amodio, Michele Di Maio, & Patrizio Piraino

The objective of this project is to study the relationship between market power and (lack of) competitiveness in the product market and labour market outcomes in South Africa, unemployment and earnings in particular. When competition on the product market is low, firms are able to keep prices high and produce lower quantities. This results in lower levels of total production compared to high-competition markets. Less production implies less employment and less dynamism in the labor market. A few very recent studies have developed new methodologies to estimate the degree of competition in different sectors and regions, and show that in the US they are highly related to labor market outcomes. We propose to carry out a similar analysis in South Africa where low competition and poor labor market performance are both very salient. Our analysis would shed light on variation across sectors and regions in employment and earnings, thus adding a new element to descriptive analyses of economic inequality in the country. For this purpose, we propose to combine the firm-level data from the South African Revenue Service and National Treasury (SARS-NT) from 2008 to 2014 with Labor Force Survey (LFS) and Quarterly Labor Force Survey (QLFS) data from 2007 to 2014. We will use the firm-level dataset to estimate markups across sectors using state-of-the art techniques in the literature. We will then use 2007 LFS data to derive a measure of baseline employment concentration in high markup sectors for each district municipality in South Africa. The combination of these two measures with QLFS data will allow us to answer the following question: do labour market outcomes differ systematically in those district municipalities where employment is more concentrated in high markup sectors? In particular, is the unemployment level higher in those district municipalities? Are there differences in labour market flows in and out of unemployment in those districts?